Hi, I am looking for a low cost weight sensor. I am trying to provide a indication of the amount of paper available. The empty container is approx. 1kg, and approx. 10kg full of paper. It should be able to handle industrial temperature range. A resolution of 1kg would be acceptable. If I can get 100g resolution, it would be great. I would prefer SPI or I2C interface, but analog out would be acceptable - something that can directly interface with a MCU. Regards Anton Erasmus
Low cost weight sensor ?
Started by ●May 22, 2006
Reply by ●May 22, 20062006-05-22
Anton Erasmus wrote:> Hi, > > I am looking for a low cost weight sensor. I am trying to provide a > indication of the amount of paper available. The empty container is > approx. 1kg, and approx. 10kg full of paper. It should be able to > handle industrial temperature range. A resolution of 1kg would be > acceptable. If I can get 100g resolution, it would be great. > I would prefer SPI or I2C interface, but analog out would be > acceptable - something that can directly interface with a MCU.It looks like one of those problems that cost $300 for a few or $.30 for 100k, but not a lot in between. Load cells would work, but will be quite expensive. Springs, levers and opto or hall sensors would be another possibility.
Reply by ●May 22, 20062006-05-22
Anton Erasmus wrote:> Hi, > > I am looking for a low cost weight sensor. I am trying to provide a > indication of the amount of paper available. The empty container is > approx. 1kg, and approx. 10kg full of paper. It should be able to > handle industrial temperature range. A resolution of 1kg would be > acceptable. If I can get 100g resolution, it would be great. > I would prefer SPI or I2C interface, but analog out would be > acceptable - something that can directly interface with a MCU. > > Regards > Anton ErasmusWhat's your volumes, and what's your willingness to build your own? As Jim Stewart mentioned there are possibilities, but the only commercially available weight (or force) sensors that I know of are all stainless steel devices meant to go into an industrial setting and last forever while being sprayed with salt water and whacked with hammers. That's good for the guy who needs one and doesn't want to spend a lot of engineering time qualifying the sensor, but not too good for a product. Some roll-your-own possibilities: Make a balance with an angle detector. The balance part is obvious, the angle detector could be a potentiometer or an LVDT or a fine encoder. Make your own load cell with your own LVDT or strain gauge. There will be qualification issues, but I suspect that most really inexpensive scales out there (like bathroom and postal scales) use this scheme. The nice thing is that LVDTs are simple in their 1st-order approximation, and a load cell can be as easy as some spring steel that's been laser cut in a certain way. Strain gauges can be harder to implement, but there may be some possibilities there, too. Use a big, compliant spring with a potentiometer or encoder. This is the same theory as a load cell, but it makes for an easier sensor design. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com Posting from Google? See http://cfaj.freeshell.org/google/ "Applied Control Theory for Embedded Systems" came out in April. See details at http://www.wescottdesign.com/actfes/actfes.html
Reply by ●May 22, 20062006-05-22
Hello Tim,> Some roll-your-own possibilities: > > Make a balance with an angle detector. The balance part is obvious, the > angle detector could be a potentiometer or an LVDT or a fine encoder. > > Make your own load cell with your own LVDT or strain gauge. There will > be qualification issues, but I suspect that most really inexpensive > scales out there (like bathroom and postal scales) use this scheme. The > nice thing is that LVDTs are simple in their 1st-order approximation, > and a load cell can be as easy as some spring steel that's been laser > cut in a certain way. Strain gauges can be harder to implement, but > there may be some possibilities there, too. > > Use a big, compliant spring with a potentiometer or encoder. This is > the same theory as a load cell, but it makes for an easier sensor design. >With the compliant spring version I'd look into capacitive sensing. Isolated plate pressed towards a grounded metal surface as spring is depressed, or the other way around. It's non-linear but a uC would likely be needed anyway in this setup. Whatever is done, this field might be peppered with patents. Regards, Joerg http://www.analogconsultants.com
Reply by ●May 23, 20062006-05-23
On Mon, 22 May 2006 12:40:12 -0700, Jim Stewart <jstewart@jkmicro.com> wrote:>Anton Erasmus wrote: > >> Hi, >> >> I am looking for a low cost weight sensor. I am trying to provide a >> indication of the amount of paper available. The empty container is >> approx. 1kg, and approx. 10kg full of paper. It should be able to >> handle industrial temperature range. A resolution of 1kg would be >> acceptable. If I can get 100g resolution, it would be great. >> I would prefer SPI or I2C interface, but analog out would be >> acceptable - something that can directly interface with a MCU. > >It looks like one of those problems that cost >$300 for a few or $.30 for 100k, but not a lot >in between.It seemed that way to me from googling, but I hoped I just did not have the right "magic" search term.>Load cells would work, but will be quite expensive. >Springs, levers and opto or hall sensors would >be another possibility.I was hoping to not have to roll my own. Regards Anton Erasmus
Reply by ●May 23, 20062006-05-23
On Mon, 22 May 2006 13:24:04 -0700, Tim Wescott <tim@seemywebsite.com> wrote:>Anton Erasmus wrote: > >> Hi, >> >> I am looking for a low cost weight sensor. I am trying to provide a >> indication of the amount of paper available. The empty container is >> approx. 1kg, and approx. 10kg full of paper. It should be able to >> handle industrial temperature range. A resolution of 1kg would be >> acceptable. If I can get 100g resolution, it would be great. >> I would prefer SPI or I2C interface, but analog out would be >> acceptable - something that can directly interface with a MCU. >> >> Regards >> Anton Erasmus > >What's your volumes, and what's your willingness to build your own? As >Jim Stewart mentioned there are possibilities, but the only commercially >available weight (or force) sensors that I know of are all stainless >steel devices meant to go into an industrial setting and last forever >while being sprayed with salt water and whacked with hammers. That's >good for the guy who needs one and doesn't want to spend a lot of >engineering time qualifying the sensor, but not too good for a product.My volume is fairly low. Initially prototypes, and then 500 or so per year. Rolling my own seems to be the only option.>Some roll-your-own possibilities: > >Make a balance with an angle detector. The balance part is obvious, the >angle detector could be a potentiometer or an LVDT or a fine encoder.This looks like it can be quite simple witjout the need for complex or time consuming calibration.>Make your own load cell with your own LVDT or strain gauge. There will >be qualification issues, but I suspect that most really inexpensive >scales out there (like bathroom and postal scales) use this scheme. The >nice thing is that LVDTs are simple in their 1st-order approximation, >and a load cell can be as easy as some spring steel that's been laser >cut in a certain way. Strain gauges can be harder to implement, but >there may be some possibilities there, too.We have used this sort of setup, but for the current product I would want something in the US$20 range.>Use a big, compliant spring with a potentiometer or encoder. This is >the same theory as a load cell, but it makes for an easier sensor design.A helix spring with a sliding potentiometer ? What sort of lengths / sizes of liding potentiometers are avialable ? Regards Anton Erasmus
Reply by ●May 23, 20062006-05-23
On Tue, 23 May 2006 01:48:18 GMT, Joerg <notthisjoergsch@removethispacbell.net> wrote:>Hello Tim, > > >> Some roll-your-own possibilities: >> >> Make a balance with an angle detector. The balance part is obvious, the >> angle detector could be a potentiometer or an LVDT or a fine encoder. >> >> Make your own load cell with your own LVDT or strain gauge. There will >> be qualification issues, but I suspect that most really inexpensive >> scales out there (like bathroom and postal scales) use this scheme. The >> nice thing is that LVDTs are simple in their 1st-order approximation, >> and a load cell can be as easy as some spring steel that's been laser >> cut in a certain way. Strain gauges can be harder to implement, but >> there may be some possibilities there, too. >> >> Use a big, compliant spring with a potentiometer or encoder. This is >> the same theory as a load cell, but it makes for an easier sensor design. >> > >With the compliant spring version I'd look into capacitive sensing. >Isolated plate pressed towards a grounded metal surface as spring is >depressed, or the other way around. It's non-linear but a uC would >likely be needed anyway in this setup. > >Whatever is done, this field might be peppered with patents.You mean that one will easily infringe some sort of patent ? I was hoping one would find cheap MEMS based devices or something that implements a cheap robust weight sensor. Units with strange gauges already mounted seem to be quite expensive. Regards Anton Erasmus
Reply by ●May 23, 20062006-05-23
On Tue, 23 May 2006 20:03:22 +0200, the renowned Anton Erasmus <nobody@spam.prevent.net> wrote:>On Tue, 23 May 2006 01:48:18 GMT, Joerg ><notthisjoergsch@removethispacbell.net> wrote: > >>Hello Tim, >> >> >>> Some roll-your-own possibilities: >>> >>> Make a balance with an angle detector. The balance part is obvious, the >>> angle detector could be a potentiometer or an LVDT or a fine encoder. >>> >>> Make your own load cell with your own LVDT or strain gauge. There will >>> be qualification issues, but I suspect that most really inexpensive >>> scales out there (like bathroom and postal scales) use this scheme. The >>> nice thing is that LVDTs are simple in their 1st-order approximation, >>> and a load cell can be as easy as some spring steel that's been laser >>> cut in a certain way. Strain gauges can be harder to implement, but >>> there may be some possibilities there, too. >>> >>> Use a big, compliant spring with a potentiometer or encoder. This is >>> the same theory as a load cell, but it makes for an easier sensor design. >>> >> >>With the compliant spring version I'd look into capacitive sensing. >>Isolated plate pressed towards a grounded metal surface as spring is >>depressed, or the other way around. It's non-linear but a uC would >>likely be needed anyway in this setup. >> >>Whatever is done, this field might be peppered with patents. > >You mean that one will easily infringe some sort of patent ? > >I was hoping one would find cheap MEMS based devices or something >that implements a cheap robust weight sensor. Units with strange >gauges already mounted seem to be quite expensive.Strain gauges are even more expensive.. Best regards, Spehro Pefhany -- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com
Reply by ●May 23, 20062006-05-23
Hello Anton,>>Whatever is done, this field might be peppered with patents. > > You mean that one will easily infringe some sort of patent ? >Possibly. I have seen some rather bizarre patents in that field. Patents where I had no clue how on earth they could ever have been granted.> I was hoping one would find cheap MEMS based devices or something > that implements a cheap robust weight sensor. Units with strange > gauges already mounted seem to be quite expensive. >Sure, if money is not a concern. But if you want to produce oodles of these and it has to be cheap you'll have to be inventive and design it around nickel and dime parts. Of course then the chances are high someone else had had the same ideas and filed for a patent. A resolution of 10% or even 1% is quite easy with a capacitive plate sensor and compliant springs. I don't know if that's patented though. Regards, Joerg http://www.analogconsultants.com
Reply by ●May 23, 20062006-05-23
Hello Anton,> >>Load cells would work, but will be quite expensive. >>Springs, levers and opto or hall sensors would >>be another possibility. > > I was hoping to not have to roll my own. >Just an idea: A couple of years ago I bought fancy bathroom scales. Nice glass design and four pillars. There must have been four pressure sensors in the pillars since wires went there and there was neither any mechanical cantilevering nor did this thing move down at all if I stepped on it. I can't look though and don't remember the brand. I returned it because it was less accurate than the old scales but it might have been good enough for you. Look at the stores (Taregt, Walmart, Mervyns etc.) I bet they still have something like that. This model was $34.99 with a big LCD display so the sensors must have been cheap. Regards, Joerg http://www.analogconsultants.com
